本文選題：小波變換 + 異構(gòu)并行計(jì)算��； 參考：《西安電子科技大學(xué)》2013年碩士論文

【摘要】：目前，隨著計(jì)算機(jī)硬件技術(shù)的飛速發(fā)展，多核中央處理器(Central ProcessingUnit, CPU)和通用計(jì)算的圖形處理器(Graphic Processing Unit, GPU)更新?lián)Q代層出不窮，但軟件應(yīng)用卻跟不上其硬件的發(fā)展速度，導(dǎo)致大量的硬件資源浪費(fèi)。為了發(fā)揮其各自的性能特點(diǎn)和提高資源的利用率，通過(guò)并行計(jì)算對(duì)密集型計(jì)算任務(wù)加速，具有很高的應(yīng)用價(jià)值和實(shí)際意義。 小波變換以其在時(shí)域和頻域都具有良好的局部化特性使得其在圖像壓縮編碼領(lǐng)域得到廣泛的研究和應(yīng)用。而隨著圖像信息數(shù)量的劇增，，對(duì)小波變換的加速處理進(jìn)而提升圖像壓縮的速度具有重要的實(shí)際意義。 本文詳細(xì)分析了小波變換的原理及過(guò)程，并闡述了CPU-GPU異構(gòu)并行計(jì)算的特點(diǎn)。在CPU實(shí)現(xiàn)小波變換基礎(chǔ)上，研究了CPU-GPU異構(gòu)并行計(jì)算的程序任務(wù)的劃分，性能的評(píng)估以及性能優(yōu)化策略。通過(guò)對(duì)小波變換實(shí)現(xiàn)的具體分析和研究，采用了基于OpenMP-CUDA混合編程模型實(shí)現(xiàn)了異構(gòu)系統(tǒng)的協(xié)同運(yùn)算。重點(diǎn)研究了9/7小波濾波在GPU上的優(yōu)化，通過(guò)對(duì)小波變換算法的優(yōu)化，使其更適應(yīng)GPU上硬件資源結(jié)構(gòu)和更能發(fā)揮并行計(jì)算性能上的優(yōu)勢(shì)。針對(duì)小波濾波中數(shù)據(jù)劃分帶來(lái)低通和高通系數(shù)相關(guān)性問(wèn)題，提出在代價(jià)最小的情況下實(shí)現(xiàn)數(shù)據(jù)的并行劃分，極大提高了程序的運(yùn)行效率�；贑PU-GPU異構(gòu)并行計(jì)算的小波變換相比CPU串行的處理速度提升了15-36倍。
[Abstract]:At present, with the rapid development of computer hardware technology, multi-core CPU Central processing Unit (CPU) and General Computing Graphics processor (GPU) are constantly emerging, but the application of software can not keep up with the development of its hardware. Lead to a lot of waste of hardware resources. In order to give full play to their respective performance characteristics and improve the utilization of resources, intensive computing tasks are accelerated through parallel computing. Wavelet transform has been widely studied and applied in the field of image compression because of its good localization in time domain and frequency domain. With the rapid increase of the number of image information, it is of great practical significance to accelerate the processing of wavelet transform and then accelerate the speed of image compression. In this paper, the principle and process of wavelet transform are analyzed in detail, and the characteristics of CPU-GPU heterogeneous parallel computing are expounded. Based on the wavelet transform implemented by CPU, the task partition, performance evaluation and performance optimization strategy of CPU-GPU heterogeneous parallel computing are studied. By analyzing and studying the realization of wavelet transform, a hybrid programming model based on OpenMP-CUDA is used to realize the cooperative operation of heterogeneous systems. This paper focuses on the optimization of 9 / 7 wavelet filtering on GPU. By optimizing the wavelet transform algorithm, it is more suitable for the hardware resource structure on GPU and can play the advantage of parallel computing performance. In order to solve the problem of correlation between low pass and high pass coefficients brought by data partition in wavelet filtering, the parallel partition of data is proposed in the case of minimum cost, which greatly improves the running efficiency of the program. The processing speed of wavelet transform based on CPU-GPU isomerous parallel computing is 15-36 times faster than that of CPU.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TP338.6;TP391.41

【參考文獻(xiàn)】

相關(guān)會(huì)議論文 前1條

1 袁林;張國(guó)峰;戴樹(shù)嶺;;基于小波變換的圖像壓縮算法研究[A];第五屆全國(guó)仿真器學(xué)術(shù)會(huì)論文集[C];2004年

本文編號(hào)：1981073